A high-throughput method for detection of DNA in chloroplasts using flow cytometry

Abstract

Background: The amount of DNA in the chloroplasts of some plant species has been shown recently to decline dramatically during leaf development. A high-throughput method of DNA detection in chloroplasts is now needed in order to facilitate the further investigation of this process using large numbers of tissue samples.

Results: The DNA-binding fluorophores 4',6-diamidino-2-phenylindole (DAPI), SYBR Green I (SG), SYTO 42, and SYTO 45 were assessed for their utility in flow cytometric analysis of DNA in Arabidopsis chloroplasts. Fluorescence microscopy and real-time quantitative PCR (qPCR) were used to validate flow cytometry data. We found neither DAPI nor SYTO 45 suitable for flow cytometric analysis of chloroplast DNA (cpDNA) content, but did find changes in cpDNA content during development by flow cytometry using SG and SYTO 42. The latter dye provided more sensitive detection, and the results were similar to those from the fluorescence microscopic analysis. Differences in SYTO 42 fluorescence were found to correlate with differences in cpDNA content as determined by qPCR using three primer sets widely spaced across the chloroplast genome, suggesting that the whole genome undergoes copy number reduction during development, rather than selective reduction/degradation of subgenomic regions.

Conclusion: Flow cytometric analysis of chloroplasts stained with SYTO 42 is a high-throughput method suitable for determining changes in cpDNA content during development and for sorting chloroplasts on the basis of DNA content.

Figure 1

Flow cytometric analysis of DAPI-stained chloroplasts. Chloroplasts isolated from seedlings at 14 (A) and 20 (B) days after imbibition were fixed in glutaraldehyde before being treated with and without DNase and stained with DAPI. The number of chloroplasts analyzed in (A) was 8800 for DNase-treated and 9000 for untreated samples. The corresponding numbers in (B) were 23,300 and 25,300.

Figure 2

Comparison of SG and DAPI using fluorescence microscopy. Chloroplasts isolated from seedlings at 20 days after imbibition stained with DAPI (B) or SG (D). (A, C) Brightfield images of the chloroplasts shown in (B) and (D). The contrast has been enhanced (inset in (B)) to accentuate fluorescence from the chloroplasts in the lower left quadrant. Scale bar is 10 μm. The exposure time for both (B) and (D) was 0.5 s. DAPI-stained chloroplasts were fixed in glutaraldehyde and SG-stained chloroplasts were not fixed.

Figure 3

Flow cytometric analysis of SG-stained chloroplasts. (A-F) Chloroplasts isolated from 20-day-old seedlings treated with and without DNase and stained with the indicated concentrations of SG. Differences in cpDNA content (G) and PPoD (H) for chloroplasts isolated from seedlings at 14 and 20 days after imbibition. (A-F) Chloroplasts were fixed in glutaraldehyde. (G and H) Chloroplasts were not fixed. At least 6000 chloroplasts were analyzed for all samples.

Figure 4

Fluorescence microscopy and flow cytometric analysis of SYTO 42-stained chloroplasts. Brightfield (A) and fluorescence (B) microscopic images of chloroplasts isolated from 14-day-old seedlings after staining with 25 μM SYTO 42. The exposure time in (B) was 0.1 s. (C-E) Flow cytometric analysis of chloroplasts from 14-day-old seedlings treated with and without DNase and stained with the indicated concentrations of SYTO 42. Brightfield (F, H) and fluorescence microscopy (G, I) of chloroplasts from immature leaves of 35-day-old plants after treatment with RNase (F, G) and DNase (H, I) and staining with 25 μM SYTO 42. Insets in (F) and (G) show brightfield and fluorescence images of a chloroplast from a different microscopic field of the same sample. The exposure times were 0.1 s (G) and 0.3 s (I). The number of DNase-treated chloroplasts that did not have visible nucleoids was 8 out of 8. The number of chloroplasts that did not have visible nucleoids was 1 out of 9 after RNase treatment, and 2 out of 19 for untreated controls. (J) Flow cytometric analysis comparing untreated chloroplasts with chloroplasts treated with RNase, DNase, or DNase and RNase from immature leaves of 35-day-old plants. (K) Comparison of unfixed chloroplasts to chloroplasts fixed with 0.8% glutaraldehyde. Differences in cpDNA content (L) and PPoD (M) for chloroplasts isolated from seedlings at 14 and 20 days after imbibition. (N) PCR amplification of a 156-bp fragment of the psbA gene from lysates prepared from chloroplasts that had been previously treated or not treated with DNase. Scale bars are 10 μm. (A, B, L, M (14-day profile)) Chloroplasts were not fixed. (C-K, L, M (20-day profile)) Chloroplasts were fixed in glutaraldehyde. (C-E, J-M) At least 2500 chloroplasts were analyzed for all samples.

Figure 5

Fluorescence microscopy and flow cytometric analysis of SYTO 45-stained chloroplasts. Brightfield (A) and fluorescence (B) microscopic images of chloroplasts isolated from 14-day-old seedlings after staining with 20 μM SYTO 45. The exposure time for (B) was 0.1 s. (C) Flow cytometric analysis of the same chloroplasts stained with the indicated concentrations of SYTO 45. Scale bar is 10 μm. Chloroplasts in all panels were not fixed.

Figure 6

Analysis of changes in cpDNA content during development using fluorescence microscopy and flow cytometry. (A-D) The frequency of relative fluorescence values of DAPI-stained, glutaraldehyde-fixed chloroplasts isolated from four different tissues at two stages of growth measured by fluorescence microscopy. Flow cytometric analysis (E) and PPoD (F) of the same four samples shown in (A-D) using SYTO 42 (chloroplasts were not fixed). D12 L1,2: first two leaves of 12-day-old seedlings. D12 Cotyledons: cotyledons of 12-day-old seedlings. D23 L5: fifth leaf of 23-day-old plant. D23 L1: first leaf of 23-day-old plant. The means (arrows) ± standard error are 6.14 ± 0.76 (A), 4.39 ± 0.81 (B), 2.62 ± 0.2 (C), 2.14 ± 0.18 (D). The numbers of chloroplasts with no detectable DNA (and the number of chloroplasts assayed) by DAPI-staining are 0(37), 0(36), 0(62), 4(66) for the samples shown in (A-D), respectively. The means of the SYTO 42 fluorescence profiles shown in (E) are 5835 (D12 L1,2), 4128 (D12 Cotyledons), 2344 (D23 L5), and 2042 (D23 L1). At least 3000 chloroplasts were analyzed. Mean SYTO 42 fluorescence values were 7607, 6007, 3115, 2431 and 8471, 5440, 3523, 2549 for the same tissues obtained from the ecotypes Nossen and Estland, respectively (profiles not shown).

Figure 7

Analysis of cpDNA content by qPCR of chloroplasts obtained from fractions spanning the distribution of SYTO 42 fluorescence values. (A) SYTO 42 fluorescence profile of chloroplasts from mature leaves of 43-day-old plants showing the four fractions collected by FACS. Chloroplasts were not fixed. (B) Schematic diagram of the chloroplast genome showing the location of the primer sets used for qPCR analysis. Numbers indicate distance (in kb) on the genome map [30] (C) qPCR analysis of cpDNA content for the four fractions described in (A). Values shown are means ± standard error for 5 or 6 replicates.